Скачать или смотреть Ethics Sheets for AI Tasks and a Case Study for Automatic Emotion Recognition

On July 15th 2021, Dr. Saif M. Mohammad, Senior Research Scientist at the National Research Council Canada presented this Language Science Talks with host UBC Computer Science Assistant Professor of Teaching Varada Kolhatkar.

Abstract: As NLP and ML systems become more ubiquitous, their broad societal impacts are receiving more scrutiny than ever before. Several high-profile events have highlighted how technology will often lead to more adverse outcomes for those that are already marginalized. This raises some uncomfortable questions for us as researchers: What are the hidden assumptions in our research? What are the unsaid implications of our choices? Are we perpetuating and amplifying inequities or are we striking at the barriers to opportunity? The answers are often complex and multifaceted. In this talk, I will make a case for continued efforts in documenting ethical considerations for AI Tasks (through individual and community efforts). I will present a new form of such an effort: Ethics Sheets for AI Tasks which, together with Data Sheets for Datasets and Model Cards for AI systems, aids in the development and deployment of responsible AI systems. Finally, I will provide an example ethics sheet for automatic emotion recognition and sentiment analysis.

I will start the talk with a quick overview of my past work at the intersection of language and emotions; notably, work on large human-annotated word–emotion lexicons.